Kyumin Hwang

Hi, I'm Kyumin Hwang, a Postdoctoral Researcher at KAIST, working with Prof. Sunghoon Im. I received my Ph.D. degree from DGIST, where I was a member of the KAIST Robotics and Vision Lab, advised by Prof. Sunghoon Im. I earned my B.S. degree from the Department of Computer Science and Engineering, Kyungpook National University, in 2020.

My research centers on multi-view stereo for 3D reconstruction and understanding, and deep learning with geometry for autonomous agents. Building on this foundation, I have recently expanded toward Physical AI, exploring how embodied agents can perceive, reason about, and act within the real world.

My long-term goal is to build intelligent agents that perceive and understand the 3D world as humans do, and go further to enable sophisticated decision-making and control for complex robotic tasks.

kyumin [at] kaist.ac.kr / Google Scholar / Github / Curriculum Vitae

News

2026

2026-08: Started as a Postdoctoral Researcher at the 5D AI Robotics Research Center, KAIST.

2026-03: One paper accepted to IEEE SPL.

2026-02: Two paper accepted to CVPR 2026.

2025

2025-11: One paper accepted to IEEE RA-L and will be presented at ICRA 2026.

2025-11: One paper accepted to AAAI 2026 as an oral paper.

2025-09: I am going to work as Visiting Scholar at Jun Gao's Lab in University of Michigan for five months.

2025-06: I received the DGIST Post-Graduate Research Abroad Award.

2025-01: One paper accepted to ICLR 2025.

2024

2024-12: One paper accepted to AAAI 2025.

2024-11: Our team received the Top Award at the 16th Electronics Newspaper ICT Paper Contest.

2024-04: One paper accepted to IEEE T-PAMI.

Publications

* indicates equal contribution
† indicates corresponding author

	CascadeOcc: Rethinking 3D Occupancy World Models with Cascaded VQ Representations Kyumin Hwang, Wonhyeok Choi, Jaeyeul Kim, Jihun Park, Daehee Park^†, and Sunghoon Im^† IEEE Signal Processing Letters (SPL), 2026 [Paper][Code]
	TaskForce: Cooperative Multi-agent Reinforcement Learning for Multi-task Optimization Wonhyeok Choi, Kyumin Hwang, Jihun Park, Kyoungmin Lee, Seunghun Lee, Jaeyeul Kim, Minwoo Choi, and Sunghoon Im^† IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026 [Paper]
	A Training-Free Style-Personalization via SVD-Based Feature Decomposition Kyoungmin Lee, Jihun Park, Jongmin Gim, Wonhyeok Choi, Kyumin Hwang, Jaeyeul Kim, and Sunghoon Im^† IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)*, 2026 [Paper]
	Scale-invariant and View-relational Representation Learning for Full Surround Monocular Depth Kyumin Hwang, Wonhyeok Choi, Kiljoon Han, Wonjoon Choi, Minwoo Choi, Yongcheon Na, Minwoo Park, Sunghoon Im^† IEEE Robotics and Automation Letters (RA-L), 2026 IEEE International Conference on Robotics and Automation (ICRA), 2026 Collaborated with the Hyundai Motor Group [Paper]
	Infinite-Story: A Training-Free Consistent Text-to-Image Generation (Oral) Jihun Park, Kyoungmin Lee, Jongmin Gim, Hyeonseo Jo, Minseok Oh, Wonhyeok Choi, Kyumin Hwang, Jaeyeul Kim, Minwoo Choi, Sunghoon Im^† The Fortieth Annual AAAI Conference on Artificial Intelligence (AAAI)*, 2026 [Paper][Project page]
	Self-supervised Monocular Depth Estimation Robust to Reflective Surface Leveraged by Triplet Mining Wonhyeok Choi, Kyumin Hwang, Wei Peng, Minwoo Choi, Sunghoon Im^† International Conference on Learning Representations (ICLR), 2025 Top Award in ICT Paper Contest [Paper][Openreview]
	Intrinsic Image Decomposition for Robust Self-supervised Monocular Depth Estimation on Reflective Surfaces Wonhyeok Choi, Kyumin Hwang, Minwoo Choi, Kiljoon Han, Wonjoon Choi, Mingyu Shin, Sunghoon Im^† The Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI), 2025 [Paper]
	A Study on the Generality of Neural Network Structures for Monocular Depth Estimation Jinwoo Bae, Kyumin Hwang, Sunghoon Im^† IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2024 [Paper][Code]

Education

	Ph.D. in Electrical Engineering and Computer Science DGIST, Daegu, South Korea Advisor: Prof. Sunghoon Im Aug. 2026
	M.S. in Information and Communication Engineering DGIST, Daegu, South Korea Advisor: Prof. Sunghoon Im Feb. 2022
	B.S. in Computer Science and Engineering Kyungpook National University, Daegu, South Korea Feb. 2020

Experience

	Post-doctoral Researcher 5D AI Robotics Research Center, KAIST, Daejeon, South Korea Advisor: Prof. Sunghoon Im Aug. 2026 – present
	Visiting Scholar Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI, USA Advisor: Prof. Jun Gao Sep. 2025 – Feb. 2026
	Research Intern Electronics and Telecommunications Research Institute (ETRI), Daegu, South Korea Advisor: Dr. Kwangju Kim Jan. 2019 – Feb. 2019, Jul. 2019– Aug. 2019

Academic Activities

Honors and Awards

DGIST Post-Graduate Research Abroad Awards ($9,500), DGIST, Jul 2025

Top Award, 16th ICT Paper Competition and Grand Exhibition, Electronic News, Dec 2024

Experiences

Invited Speaker of 5D AI Robotics Initiative (5D AI-RI) InnoCORE Workshop, KAIST, Jul 2026

Teaching Assistant of Advanced Deep Learning, DGIST, Spring 2022

Teaching Assistant of Introduction to Deep Learning, DGIST, Fall 2020

Reviewer

International Conference on Learning Representations (ICLR)

Association for the Advancement of Artificial Intelligence (AAAI)

Neural Information Processing Systems (NeurIPS)

European Conference on Computer Vision (ECCV)

IEEE Robotics and Automation Letters (RA-L)

IEEE Signal Processing Letters (SPL)

Website Template